What is Gantry?
Gantry helps you figure out how your models are performing in production and how you can improve that performance. We do this by capturing the data going through your models at every stage of the process and allowing you to make sense it via our analysis and visualization tools.
Analytics to improve your models
Log inputs, outputs, predictions, and feedback from your ML-powered product by adding a few lines of code.
Configure performance indicators for your models and establish baselines with analytics powered by our SDK and dashboard. Use our SDK and dashboard to discover user cohorts, data slices, and edge cases where your model is biased or underperforms in production.
In Gantry, every dashboard query is also available in our Pandas-like Python SDK , so you can continue to explore all of your production data in a Jupyter notebook.
Make sense of unstructured data
Uncover a wide range of patterns in your text, images, or other messy data without writing code using our off-the-shelf Projections.
You can also extend Gantry with your own custom projections to build a monitoring and data analysis suite unique to your model.
Detect concerning model behavior before it becomes a problem
Gantry allows you to compute performance indicators about data quality, data drift, model performance,
and user satisfaction.
You can set thresholds on those performance indicators to receive notifications, in Slack or via a generic web hook, about degradations. Easily configure daily or realtime notifications with just a few clicks.
How is Gantry different?
- Scalability. Gantry is built on cloud-native technologies and designed to scale to terabytes of data.
- Flexibility. Uncover a wide range of patterns in your text, images, or other messy data without writing code using our off-the-shelf projections.
- Extensibility. Extend Gantry with your own custom projections to build a monitoring and data analysis suite for your unique model.
- Security. Gantry is built with enterprise-grade authentication. We’re SOC 2 compliant.
Updated 22 days ago